Comparing dissolution profiles using the similarity factor (f2)
Biorelevant generic dissolution tests are performed to examine whether the test product formulation has the same release profile as the reference drug.
The similarity factor (f2) calculation is performed to quantitatively assess the similarity between two dissolution profiles, by using the difference in % release at each time point, as calculated below:
Where R and T refer to the % release of reference and test product respectively, t represents a given time point, and n represents the number of time points measured. The conditions for using the f2 are as follows;
• The f2 shouldn’t be used if both profiles reach 85% release within 15 minutes
• Only one measurement should be used after both profiles reach 85%
• The relative standard deviation (RSD%) within each data set should be ≤20% up to 15 minutes and ≤10% after 15 minutes
• The FDA states that bioequivalence is reached when f2 ≥ 50, and this value is required to grant a biowaiver (FDA, 2017)
There is one main issue with the f2; it calculates similarity relative to the 0 to 100% bounds, irrespective of the maximum solubility of the dosage forms. Graph A and B (below) are pairs of profiles where the test product exhibits 50% less release than the reference at all time points (i.e. same degree of difference relative to maximum solubility).
Max. solubility = 100% Max. solubility = 50%
Despite both pairs of profiles having the same relationship, the results show that the Rt – Tt differences values are greater in A than in B, giving a lower f2 value. This demonstrates that doses with higher maximum solubilities are more susceptible to generating greater differences, which gives misleading results.
Another point to note is that the f2 calculation is sensitive to the number of measurements. To get an accurate comparison for different f2 values, the timepoints that are measured need to be standardised.
In conclusion, the f2 value is convenient, rather than accurate; the results it generates can be misleading, especially when comparing dosage forms with different maximum solubilities. Nonetheless, it is still useful as a rough and ready method of comparing two dissolution profiles.
Food and Drug Administration, CDER (December 2017). Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate-Release Solid Roal Dosage Forms Based on a Biopharmaceutics Classification System.